In response to RFA-AI-14-071, we propose to determine the utility of assay of tenofovir-diphosphate (TFV-DP) in dried blood spots (DBS) as an objective measure of adherence to antiretroviral treatment (ART) in South Africa. Measurement of this active antiretroviral medication (ARV) anabolite quantifies medication ingestion, may provide early warning of viremia, and may be used to provide motivating and actionable feedback to patients and providers. Feedback that motivates adherence has additional importance as more patients begin ART earlier in their disease course when they are feeling healthy and, as a result, may have problems sustaining long-term adherence. Routine (often only annual) viral load measurement often misses the opportunity to prevent viremia and/or development of resistant virus. Objective adherence measures could provide timely feedback for earlier intervention, however, existing measures have problems with validity; patient/provider burden; and expense and acceptability. None indicates if a dose is actually ingested. Co-Investigator, Anderson, developed an assay of TFV-DP in DBS that reflects drug ingestion over past weeks/months; research likely to lead to a point- of-care application of this assay is under development. Although the assay has been characterized in HIV- adults, our proposed project would examine its utility for treatment rather than prevention in real-world settings in Sub Saharan Africa. This research will provide critical information without which the effective roll-out of such technologies would be significantly compromised - especially in low-resource settings, such as South Africa. Following our recent pilot study (Remien et al.) using Anderson's DBS TFV-DP assay among HIV+ patients in SA, we propose research on the use of this assay to assess and manage ART adherence in this low-resource setting through two Specific Aims. Aim 1: Among 250 HIV+ adults on ART containing TFV for ?12 months, we will determine the ability of this assay to provide an objective, clinically relevant, and actionable measure of adherence by using monthly drug anabolite and VL assays along with continuous Wisepill output over 12 months. We will compare the abilities of the DBS TFV-DP assay and Wisepill to predict viremia; determine the magnitude of decrease in DBS TFV-DP - in real-world use - that predicts viremia and by how long; and document the range of DBS TFV-DP levels in patients who remain virally suppressed. Aim 2: We will develop messages and procedures for giving patients and providers monthly feedback from the assay and, in a small pilot study (N=60), examine provider and patient behaviors in response to receiving this information. Findings will inform the field and provide pilot data for a future larger trial to establish the imact of such feedback on patients' adherence-related behaviors and on medical management by providers. The proposed research represents an ongoing, innovative, and productive multidisciplinary partnership among behavioral scientists, pharmacologists, and clinicians in the US and South Africa.